What is the relationship between ethnic diversity, political institutions, and public goods provision? This is a salient topic that has been much studied. However, there is no consensus among scholars: some find ethnically heterogeneous countries have higher level of public goods provision (e.g. Switzerland) while others do not; and some proportional representation (PR) electoral systems can induce stability of the political system while others do not. Joel Sawat Selway's Coalitions of the Well-being: How Electoral Rules and Ethnic Politics Shape Health Policy in Developing Countries tries to answer this puzzle by examining the interactive effect of ethnic fractionalization, ethno-income cross-cuttingness, ethno-geographic distribution, and electoral rules on provisions of a critical public good: national health policy.
Selway begins his analysis with a question: Under what conditions do politicians adopt policies offering public goods? Easterly and Levine argue high ethnic fractionalization can lead to a low provision of public goods; Lijphart contends proportional electoral rules can stimulate welfare-oriented outcomes; and Horowitz claims the alternative vote system can motivate inter-ethnic coordination. Selway challenges these conventional wisdoms and emphasizes two points: ethnic fractionalization per se is not deterministic in terms of public goods provision, and there is no one-size-fits-all electoral system for divided societies. In low ethnic-salient countries, proportional or mixed systems provide politicians with more incentives to attract broader constituencies than the first-past-the-post (FPTP) rule. In countries with a high degree of ethnic fractionalization and low degree of ethno-income cross-cuttingness, FPTP is better able to induce inter-ethnic coalition and public goods provision than the PR system if ethnic groups are geographically intermixed. This is because no one ethnic group can single-handedly win the majority of parliamentary seats. In contrast, if a country has high ethnic saliency and the groups are geographically isolated, neither the majoritarian nor the proportional system can induce public good provisions because elections simply “translate societal diversity into the legislature” (p. 213).
Selway employs both quantitative and qualitative methods to test these arguments—rendering the findings in this book powerful. Not only do the statistical models verify Selway's hypotheses, but the case studies further elucidate the mechanism linking electoral system designs and health policies (and outcomes) in low ethnic-salient countries, high ethnic-salient countries with geographic intermixing ethnic groups, and high ethnic-salient countries with geographic isolated ethnic groups, respectively. The statistical models cover a large number of countries, and the case studies deeply discuss social structure, electoral systems, parties’ or candidates’ strategy to win elections, and health policies within each selected case. Furthermore, the author's selection of cases simultaneously considers treatment and control variables. Selway compares cases having similar social structure but different institutional designs (Myanmar and Indonesia), cases having similar electoral rule but different social structures (pre-1997 Thailand and Mauritius), and even cases having similar social structure, electoral rules, and outcomes (Botswana and pre-1993 New Zealand). Selway clearly explains why these cases are comparable and why the difference and similarity between cases can be explained by his arguments.
Despite the strength of this book, there are two shortcomings. First, the methods of calculating ethno-geographic cross-cuttingness (EGC) are problematic with regards to validity. Selway generally calculates this index based on survey data. However, many survey conductors use the Probability Proportional to Size (PPS) Sampling method rather than the Simple Random Sampling method to sample respondents (in order to reduce the cost of survey project). What this means is that while respondents sampled by the PPS method are still selected randomly, the sampled geographic distribution may not be identical to the geographic distribution of the population. For example, Selway uses the World Value Survey data to calculate the EGC of Malaysia. But this survey did not include any respondent from the capital, Kuala Lumpur, whose population is greater than seven states according to the census data in 2010. To overlook the biggest city in a country is problematic.
Second, Selway neglects the effect of populism on policy making. In the Thailand case, he does not consider Thaksin's character—his business experience and skills—as an alternative explanation for the changing health policy after 1997. As such, we do not know whether Thaksin's health care policy proposal is due to the new electoral system or a populist campaign strategy. A populist politician usually raises national issues and attracts many votes from broader constituencies under any electoral rule. Consider Thailand after 2007: even though electoral rules were changed back to the block vote, the populist People's Power Party continued with the policies that had been established by the pre-coup Thai Rak Thai party under the mixed member system. As we have seen in Latin America, national health care policy can be proposed because of populist candidates rather than changes in electoral rules. Selway's argument would have been more persuasive if he had situated the Thailand comparison against a Latin American country.
In conclusion, readers will enjoy reading this book due to its powerful empirical evidence and its deep qualitative analysis of each case. This is an outstanding academic work that scholars who are interested in political institutions, ethnic politics, and health policies should not miss.
